1
From Economist to Strategist
One of the main findings of this study is the degree to which Thomas Schellingâs powerful and influential strategic thinking of the 1960s was influenced by questions he had begun to ask as an economist in the early postwar years. A second, and perhaps even more important finding, is the way that Schellingâs continuing interest in the concept of stability provides a framework for understanding his overall approach to strategy. This chapter begins the process of outlining these key themes, and shows them to be richly intertwined in the evolution of Schellingâs career.
Following a brief biographical summary, this chapter moves through Schellingâs main early works, many of which have not been highlighted at all in the existing literature which deals with him. These pieces include his very first articles in economics journals in the mid-1940s as well as later and better-known works wherein Schelling finds military applications for his ideas on bargaining. The culminating point for the chapter is Schellingâs 1960 classic, The Strategy of Conflict. This is not to deny that he continued with similar themes in his writings on a whole range of public-policy issues in the last three decades of the twentieth century, but by the early 1960s the main evolution in his strategic thinking had occurred, and a mature concept of stability can be located. Schelling himself identifies the movement of American strategic thought from the research institutes and universities to âthe Establishment, the Departments of State and Defenseâ in 1961 as a significant turning point: âThe idea stage was about over then, although books reflecting earlier thought and work continued to appear in the early 1960sâ.1
Thomas Crombie Schelling was born in 1921 at Oakland, California.2 He began his undergraduate studies before the Second World War at the University of California Berkeley, where he was introduced to economics by William Fellner. Schelling received his BA from Berkeley in 1943, where he remained to pursue some graduate studies before working in Washington for the United States Budget Bureau under another prominent economist, Arthur Smithies.3 In 1946 Schelling moved to Harvard University where his work in economics, as both student and lecturer, resulted in some early publications relating to the question of income.
Between 1948 and 1953, Schelling worked as an economist for the US government in Denmark, Paris and Washington DC. During this period he was involved in the allocation of Marshall Plan funds as a member of the Economic Cooperation Administration.4 Schelling returned to academia in 1953 as an Associate Professor of Economics at Yale University. During his five years at Yale, he produced a number of studies of foreign assistance and bargaining, including essays which were later to become part of The Strategy of Conflict.
In the summer of 1956 Schelling attended a discussion group on the east coast set up by the RAND Corporation,5 which was to play such an important role in the development of his thinking about strategy and stability. He spent the summer of 1957 at RANDâs headquarters in Santa Monica, where he was drawn increasingly into the study of military problems and to the group of analysts working on them. During this time he was invited by Charles Hitch, the Director of Economics, to return to RAND for a full year. Before Schelling took up this offer in September 1958, he spent several months in London working on game theory and became Professor of Economics and Associate of the Center for International Affairs at Harvard.6 In the event, his longer stay at RAND was perhaps the most productive and instrumental single year of his career, not least because of the stimulating environment he enjoyed in the company of other strategic thinkers who were interested in similar questions.
Back on the east coast in 1959,7 Schelling was one of the leading players of the other main stimulus for American strategic thought, the âHarvardâMIT axisâ.8 As a strategic thinker, he was particularly influential during the early and mid-1960s, publishing the often-quoted The Strategy of Conflict in 1960, Strategy and Arms Control with Morton Halperin in 1961 and Arms and Influence in 1966. The thinking behind this body of work was a major inspiration for the development of US strategic policy during this period.
Since the late-1960s, Schelling has increasingly extended his range of research to include such issues as segregation, organised crime and energy and environmental questions. In 1969, he was appointed Lucius N. Littauer Professor of Political Economy at the John F. Kennedy School of Government, Harvard University, and in 1977 he won the Frank N. Seidman Distinguished Award in political economy. Since 1990, he has been Distinguished Professor of Economics and Public Affairs at the University of Maryland, College Park.
EARLY WORK ON INCOME ECONOMICS AND BARGAINING THEORY
Schellingâs article âRaise Profits by Raising Wagesâ, which appeared in the prominent journal Econometrica in 1946, warrants mention not only because it was his first publication. It is also notable because in Schellingâs analysis of the relationship between wages, profits and national income, there are already clear signs of an interest in stability â albeit the stability of an economic system. Schelling argues that if the sum of a number of crucial propensities (to spend, to invest and so on) are greater than unity (i.e. their combined probabilities are greater than one), then the âassumption of a stable systemâ no longer holds. In such a case, âthe system is âexplosiveâ â i.e., without a finite multiplierâ.9 With this analysis Schelling had presented a basic model of an unstable or âexplosiveâ strategic situation several years before he had developed an interest in strategy.10 (Very much the same sort of model was to be applied by Schelling in understanding âexplosiveâ situations in strategy.)
In Schellingâs second publication, his main concern is the stability of economic growth (national income) which he understands in terms of the stability of dynamic equilibrium. Schelling describes the âequilibrium aspectâ as âthe tendency for the appropriate rate of growth to maintain itselfâ.11 An unstable system is thus one where the equilibrium is âunable to survive disturbancesâ,12 when the â[d]isturbances from equilibrium are self-aggravatingâ.13 Hence the search for âstabilizersâ is to seek âmore resiliency in the system, so that disturbances will be cushionedâ.14 The idea of stability in terms of self-aggravation versus self-maintenance is very important for his later strategic thinking,15 and understanding stability in terms of the ability to retain an equilibrium is discussed in later portions of this study as the âsecond aspectâ of Schellingâs stability concept â the steadiness of the equilibrium or bargain.
The question of income dominated Schellingâs research and teaching interests at Harvard. In 1951 he produced his PhD dissertation entitled âNational Income Behavior: An Introduction to Algebraic Analysisâ and his first book was published under the same name in the same year.16 In this text, Schelling develops and solves equations for several variations of national (and international) income behaviour. His main concern is the stability of these various solutions and it is once again a dynamic stability in terms of the tendency for the function to move towards â rather than away from â the given solution. This work builds on an earlier piece in which Schelling treats solutions as equilibria in a discussion of the dynamic stability of the relationship between price and unemployment levels.17
Hence, Schellingâs initial interest as an economist was the dynamic stability of income functions. But it was his interest in bargaining, developed while he was a graduate student at Harvard18 and further stimulated by his experiences as an economist in Europe, which was to brook the divide between economics and questions of strategy and provide him with more valuable insights into the stability question. Shortly before he left Harvard in 1948, Schelling read some reviews of von Neumann and Morgensternâs Theory of Games, which prompted him to work on an unpublished paper in which he criticised this pioneering text for its treatment of games involving bargaining situations.19 Schellingâs argument was that a likely solution to these sorts of games was the ability of either side in a bargaining relationship to commit or bind itself to a particular outcome.20 This approach was to typify Schellingâs subsequent writing on bargaining and game theory. His interest in this area was intensified by his experience working as an official economist when he had noticed that almost everything associated with the Marshall Plan and NATO had involved bargaining. Hence by the time he arrived at Yale, Schelling was determined to work on this subject.21
In terms of Schellingâs published work, an early sign of this interest is his argument in a paper published in 1955 that foreign assistance programmes could be analysed profitably as matters of bargaining. For instance, when many countries were involved in the provision of assistance, the process by which they settled on the division of costs between them could be seen as the resolution of a bargaining situation. Because there were many possible distributions of the costs between the countries providing funds, the main problem was finding a single outcome upon which there could be widespread agreement;
in many negotiations there is a wide area of outcomes that would be preferable, to all parties concerned, to a breakdown of negotiations. The concepts of âneed,â âcapacity,â âequity,â and so on, help fill the vacuum of indeterminacy in such negotiations. Precedent helps fill the same vacuum; and having been used, the equity concepts are even more likely to be used again.22
With this line of argument, Schelling had laid down in print the approach to bargaining which was to characterise his later strategic analysis. The key question was how to fill the âvacuum of indeterminacyâ â to locate those items (âthe procedures and the language and the symbolsâ23) which were able to âconstrain the outcomeâ.24 The idea that precedent can pr...